Practice Update: DIABETES

MY PERSPECTIVE 22

Reading our blueprints – nature’s lessons in pathophysiology found within our DNA By Allison B Goldfine MD & Alessandro Doria MD, PhD, MPH

Dr Goldfine is Associate Professor at Harvard Medical School, and Co-Head of the Section of Clinical Research at Joslin Diabetes Center in Boston. Dr Doria is Associate Professor in the Department of Epidemiology, Department of Epidemiology, Joslin Diabetes Center and Harvard Medical School, Boston.

3-Hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase inhibitors – statins – have profound beneficial effects on cardiovascular event rates but are also associated with a higher risk of incident type 2 diabetes. Whether this is attributable to low-density lipoprotein cholesterol (LDL-C) lowering, per se, or to direct or indirect off-target effects of statins remains poorly understood.

consistent with the working hypothesis, variants in NPC1L1, and to a lesser extent HMGCR and PCSK9, were also significantly associated with an increased risk of diabe- tes. A nonsignificant trend toward a similar effect was also observed for ABCG5/G8 and LDLR. These findings agree with those by Swerd- low et al, who also found an association between HMGCR and increased risk of type 2 diabetes, and those by White et al, who described a similar diabetes-pre- disposing effect with a genetic risk score based on 130 LDL-C–associated SNPs. 2,3 Taken together, these studies provide strong support for LDL-C lowering con- tributing to development of diabetes. However, whether the culprit is LDL-C low- ering, per se, or how this goal is achieved, remains unclear. Because not all the genes hosting variants associated with lower LDL-C showed statistically robust associa- tions with diabetes in the study by Lotta et al., it is possible the molecule or metabolic function targeted by the LDL-C–lowering drug matters most for development of diabetes. 1 On the other hand, the fact that

I n a study recently published in JAMA , Lotta and colleagues addressed this question by exploiting nature’s own experiments through a Mendelian rand- omization study. 1 Specifically, LDL-lowering alleles in or near the HMGCR, NPC1L1, and PCSK9 gene encoding targets of LDL-low- ering drugs (statins, ezetimibe, and PCSK9 inhibitors, respectively) and other LDL-C– related variants near the ABCG5/G8 and LDLR genes were used as proxies to assess whether associations between pharmacological LDL-C lowering and risk of diabetes is causal. In a meta-analysis of United States and European cohorts within large genetic association studies, each LDL-lowering variant was associated with a lower odds ratio for coronary artery dis- ease, with similar effect sizes per 1 mmol/L (39 mg/dL) reduction in LDL-C. However,

PRACTICEUPDATE DIABETES